1. Introduction

Cancer is a global health problem and one of leading causes of mortality worldwide in the last decades, and the primary prevention of cancers is an area of great interest from scientific, economic, and political levels [1,2]. Numerous risk factors associated with the progression of cancers including environment, genetic hallmarks and lifestyle factors such as diets, smoking, overweight and physical inactivity have been confirmed [3,4]. Due to the genomic instability and phenotypic variation during tumor progression, a potential therapeutic demand is to understand the underlying mechanisms and to drive the combinatorial interventions including drugs, diets and exercise for providing potentially great advantages with intrinsic multi-target effects during the prevention and treatments of cancers. Spermidine as one of the polyamine members is a trivalent cationic compound found in eukaryotic cells, particularly abundant in sperm. By interacting with nucleic acids, proteins, ATP and other polyanions through electrostatic binding, spermidine is indispensable in cell division and proliferation through maintaining DNA genomic homeostasis, regulating gene transcription and translation, and modulating autophagy, apoptosis, oxidative stress, angiogenesis and cell-to-cell communication [5]. Spermidine has been proven to be a precursor for the essential enzymatic modification of eIF5A, and is required for cell proliferation and viability [6]. 

Importantly, there is growing consensus that spermidine can induce autophagic flux through reducing acetylation level in cells, thereby displaying pleiotropic effects including promoting lipid metabolism, accelerating anti-inflammation, improving antioxidant activity, and enhancing mitochondrial metabolism and respiration [7–9]. Autophagy plays a crucial role in cell differentiation and tissue remodeling, ensures cellular homeostasis and proteostasis, and acts as a cell housekeeper by preventing the accumulation of damaged or toxic proteins and organelles, thereby enhancing the resistance to cellular stress during the progression of aging and diseases [10–12]. Polyamine synthesis is down-regulated in the senescent status of many tissues [13]. Consistently, the appropriate level of spermidine in vivo has been assumed to be beneficial to longevity in an autophagydependent manner. During aging process, the content of polyamines in whole blood of elderly population presents a gradually decreasing trend as the extension of age [14]. Moreover, new evidence shows that spermidine can protect from pathological events including two major death causes: cardiovascular disease (CVD) and cancer [15], and other aging-related diseases such as cognitive impairment during Alzheimer’s disease (AD) and Parkinson’s disease (PD) [16,17]. With regard to cancer, the dys-regulation of polyamine metabolism is a defined signature of many types of tumors [18]. As a caloric restriction mimetic and autophagy inducer, spermidine can reduce the growth of transplantable tumors, stimulate immune surveillance in combination with chemotherapy, and suppress tumorigenesis induced by chemical insults in mice [19].

 Dietary polyamine supplementation is correlated with lower cardiovascular disease and cancer-related mortality in human. A recent epidemiological study has documented a potential association between dietary spermidine intake and prolonged survival in human, suggesting that individuals supplied with long-term spermidine-rich diets are characterized by lower overall mortality including CVD and cancers [15]. Moreover, a protective effect of exogenous polyamines is confirmed in postmenopausal women with colorectal cancer risk-lowering behaviors such as reducing body mass index and increasing fiber intake [20]. Therefore, dietary supplementation with spermidine can be served as a promising prevention strategy in various aging-related health issues. However, based on positive regulation of cell growth and proliferation by polyamines, spermidine at too high level could be detrimental to patients suffering from cancer, aging, innate immunity and cognitive impairment during AD and PD [21]. Spermidine is already present in many foods originating primarily from raw plant and animal tissues in our diets, and the level of spermidine is profoundly affected by its external supply [22]. Thus, it is crucial to quantify spermidine level in dietary sources with safety and tolerability as an adjuvant in current standard management strategies against cancer.

2. The sources of spermidine

Spermidine is involved in many physiological processes of plants and animals. In general, a broad and diverse palette of foods, including plant and animal origin foods, contain spermidine at a high amount. 

Since spermidine can be adsorbed from dietary sources, it also can be produced by intestinal microorganisms [23]. Spermidine is the polyamine member easily absorbed from human intestine and distributed in the body without degradation through systemic circulation [24]. Thus, spermidine-rich foods can contribute to its increased concentration in multiple systems. High amounts of spermidine are detected in many vegetables and fruits including dried soybean (207 mg/kg), mushrooms (62.4–139.3 mg/kg), green peas (4.5–94.5 mg/kg), lettuce (14.8–104.1 mg/kg), broccoli (24.5–51.8 mg/kg), and mango (30 mg/kg). In grains, Japanese corn, whole grain, brown rice, and millet are also rich in spermidine [21]. Spermidine in plants such as safflower and tea has also been gained the large attention in pharmaceutical industries [25,26]. On the other hand, meat, seafood and dairy foods also have high-level spermidine [27]. Moreover, the products from fermentation processes with polyamine-generating bacteria and fungi used in the food industry can provide the generation of polyamines through microbial strains, which may contribute to the malodorous properties of milk and soybean products such as natto, amaranth grain, durian and a verity of cheeses sometimes [15,22]. Therefore, the circulating level of spermidine is influenced by consuming polyamine-rich foods directly or diets containing spermidine-producing microbiota indirectly.

 Although the decarboxylation of arginine under the catalysis from arginine decarboxylase (ADC) seems to be associated with the synthesis of agmatine in bacteria and plants, it is found to encode an ornithine decarboxylase antizyme inhibitor (AZIN), and the purified protein lacks ADC activity [31]. Then, the biosynthesized putrescine can be sequentially converted into spermidine and methylthioadenosine (MTA) by spermidine synthase (SRS) with sequential reactions of aminopropyl group from decarboxylated S-adenosylmethionine (dcSAM), which is converted from S-adenosylmethionine (SAM) by enzymatic role of adenosylmethionine decarboxylase (AdoMetDC) as a rate-limiting enzyme [32,33]. SMS can produce spermine and an additional MTA molecule from the secondary dcSAM molecule and spermidine. SAM serves as a methyl group donor in many methyltransferase reactions including histone and DNA methylation that are necessary for the epigenetic control of development and aging [34]. Moreover, spermine can be oxidized directly and specifically to produce spermidine by spermine oxidase (SMO), with the production of H2O2 and 3- aminopropanal [35]. During polyamine biosynthesis, metabolic factors can affect the bioavailability of arginine, which is important for the production of nitric oxide (NO) as an important signaling molecule in tumorigenesis [36,37]. The catabolic mechanism of spermine and spermidine is a two-step process catalyzing by rate-limiting catabolic enzymes such as spermidine/spermine N1 -acetyltransferase (SSAT) and N1 -acetylpolyamine oxidase (APAO). SSAT plays an important role in polyamine homeostasis with the conversion from spermine and spermidine to monoacetylated metabolites [38].

Different polyamines and enzyme levels in different cancers are diverse so that different cancers have different metabolisms. Therefore, it may not be optimal to define the complex relationships between cancers and elevated polyamine levels in all type of cancers. Based on higher levels of polyamines in tumors and positive regulation of cell growth and proliferation, the elevated level of spermidine has been recognized as the biomarker for monitoring the growth of tumors [53]. Since the proliferation-enhancing and cytoprotective effects of polyamines on cultured cancer cells or xenografted tumors in immunodeficient mice [54], polyamines may have pro-carcinogenic properties. It is reasonable to use the targeted metabolic pathway of polyamines for the prevention and early diagnosis of cancers, especially in characterizing different types of cancers. The increased activity of the enzymes involved in polyamine biosynthesis is also observed in tumors, especially ODC [55]. As mentioned above, ODC is overexpressed as a key rate-limiting enzyme in polyamine synthesis pathway and regulated at the transcriptional level by tumor-promoting agents in a variety of cancers [56]. ODC is the target of the oncogene Myc as a potential oncogene because it can be overexpressed in transformed mammalian cell lines alone or in combination with other oncogenes [57,58]. Transgenic mouse models overexpressing ODC in skin have demonstrated that ODC can lead to the formation of spontaneous skin tumors, suggesting its possible association with cancerous cells [59]. However, life-long overexpression of ODC gene in transgenic mice does not result in the enhanced spontaneous tumor incidence or neuronal degeneration [60]. 

Besides, polyamine oxidation through SSAT/APAO pathway is peroxisomal oxidation in the presence of peroxisomal catalase, thus substantially attenuating the production of H2O2. Because increased polyamine level is usually associated with poor prognosis, and reveals a decrease after tumor eradication and an increase after relapse, the effect and underlying mechanisms of polyamines on the metastasis and invasion of cancer cells have attracted extensive attention for further investigation [61]. In cancer tissues, polyamines reveal an increased level so that the increased uptake of spermine and spermidine may be associated with increased production of proteinases for degrading surrounding tissues [62]. Moreover, polyamines play an important role in inflammation-induced carcinogenesis, with decreased immune functions and enhanced capability of cancer cells for the invasion and metastasis to new tissues as the increased spermidine and spermine levels [61]. These findings implicate another role for polyamines in cell migration or metastasis in cancers. In recent years, selective inhibitors have been developed for regulating the metabolism of polyamines. Some inhibitors have become as the important tools in elucidating metabolic products of polyamines, but only few of them have been used as the effective inhibitors for controlling the growth of tumors in clinical trials (Table 2). A number of therapeutic practices have been conducted using difluoromethylornithine (DFMO), an inhibitor of ODC, as a chemopreventive agent based on its capability to control the remission of tumors in either animals or human with low toxicity [63,64]. In many tissues such as colonic mucosa, DFMO can suppress putrescine and spermidine,but not spermine [65,66]. 

DFMO exposure could not suppress the expression of genes involved in cell proliferation, tissue remodeling and tumor invasion in vitro in several types of tumors with decreased spermidine and putrescine levels, but can exhibit limited anti-tumor activity in vivo and in clinical trials due to compensatory mechanisms upon the occurrence of depleted polyamine pools, suggesting a novel combinatorial therapeutic strategy with polyamine-targeted drugs [67]. Several competitive inhibitors of the enzymes for polyamine biosynthesis, including methylglyoxal bis(guanylhydrazone) (MGBG) and Sadenosyl-methionine-decarboxylase (SAM486A), with the functions of decreasing spermidine and spermine levels and increasing putrescine level, have also been tested in vitro and in clinical trials for various types of tumors (Table 2). These inhibitors associated with polyamine metabolism have revealed anti-proliferative effects in cell and animal models, while their successful clinical application in human are focused on glioma, colon cancer and non-melanoma skin cancer [68]. Although it is plausible that the intervention through polyamine signal pathway may not be an effective therapeutic approach as a single agent in other solid tumors [69–72], these inhibitors still can provide a large number of preclinical data along with chemoprevention and additive anti-tumor effect when combined with other agents [68]. In general, these inhibitors are well tolerated in most trials. The treatments at high doses in these trials usually can result in ototoxicity and gastrointestinal toxicity accompanying with hearing loss, diarrhea, fatigue, headache and nausea in a significantly large number of patients (Table 2).

The dys-regulated apoptosis with overexpression of anti-apoptotic proteins and/or down-regulation of pro-apoptotic proteins often leads to the accumulation of potentially harmful cells, including cancer cells, and contributes to tumorigenesis and the resistance to anti-cancer treatments [127]. Since the suppression of apoptosis is thought to play a central role in the development and progression of some cancers, targeting critical apoptotic program is an attractive therapeutic way to induce the death of these cells. In contrast to the tumor-suppressor roles of autophagy, stress-activated autophagy may promote the survival of tumor cells, especially when apoptosis is defective [128]. The overaccumulation of polyamines is associated with cell transformation or apoptosis. Recently, increasing evidence suggests that spermidine can suppress the proliferation and promote the death of HeLa cells via autophagic activation in vitro [129]. In accordance with these results, synthetic acyl-spermidine derivatives as polyamine analogs have shown pro-apoptotic effects in human breast cancer MCF-7 cell line [130]. In addition, the activation of amine oxidation by diamine oxidase can also cause oxidative stress and apoptosis by the generation of H2O2 and reactive aldehydes [131]. Furthermore, maize polyamine oxidase can also induce the apoptosis of LoVo human colon adenocarcinoma cells by catalyzing oxidative deamination of spermine and spermidine to generate H2O2 and aldehydes [132]. The tumor repressor p53 with the capability to induce SSAT is involved in ferroptotic cell death, which also provides the insight into the regulation of polyamine catabolism and ferroptosis-mediated tumor suppression [133].